Skip to content
SkillModel-led2,289 tokens in use

writing-plans

Turns a spec into a bite-sized implementation plan with exact files, signatures and tests for an engineer with no context.

322,762 stars on the repository MIT Licenseobra/superpowers · skills/writing-plans
/plugin install superpowers@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing writing-plans costs

21tokens idle
2,289tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

gitdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure17/25
  • Examples12/15
  • Output15/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs git commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 135

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity56
  • Maintenance68
  • Community64
  • Maturity36
  • Momentum40

Writing Plans produces the plan that execution skills follow. It maps the file structure first, splits work into tasks sized around a test cycle and a reviewer’s gate, and writes steps that each carry exactly what an implementer cannot decide alone: file paths, signatures, spec values, failing tests and the commands that prove them.

What You Get

  • A plan saved under docs/superpowers/plans
  • A header with goal, architecture, spec link, global constraints and review focus
  • Tasks with files, interfaces and test-first steps ending in a commit
  • A five-point self-review
  • A handoff that recommends an execution method with one reason

Common Use Cases

  • Planning a multi-step feature from an approved design
  • Splitting a large spec into separate plans per subsystem
  • Preparing work for a cheaper model to execute

What Makes It Unique

A plan is the set of decisions the implementer cannot make alone. The skill says a step is done when the implementer can write exactly one reasonable thing from it, warns that a plan longer than the code is a transcript, and adds a review focus list of failure modes the tests would otherwise miss.

Under The Hood

Instructions only. The agent reads the spec, writes a markdown plan file into your repository and links it for review. The skill ships no scripts and calls no outside services. It names the two execution skills it hands off to.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers